Summary Reasons for performing study: The reasons why equine grass sickness (EGS) recurs on premises are unknown and, consequently, practical methods for reducing the risk of recurrence are not available. Objectives: To identify risk factors associated with recurrence of EGS on premises and to gain possible insights into the pathogenesis of the disease. Methods: Data on disease history and risk factors were collected by postal questionnaire from premises with EGS cases between 1st January 1997 and 31st December 2001. Data on variation in rates of recurrence of EGS for different risk factors were analysed using Poisson regression analysis. Results: Of 509 premises contacted, 305 (60%) returned useable questionnaires and 100 of these (33%) were classified as ‘recurrent’ premises. An overall median incidence rate for EGS of 2.1 EGS incidents/100 horses/premises/year was recorded. There was an increased rate of recurrence with higher numbers of horses, presence of younger animals, stud farms and livery/riding establishments, loam and sand soils, rearing of domestic birds and mechanical droppings removal. The rate of recurrence decreased with chalk soil, cograzing ruminants, grass cutting on pastures and removal of droppings by hand. Several statistically significant interactions were identified. Conclusions: Many of the findings are consistent with the theory that EGS is a toxico‐infectious form of botulism. Several of the significant factors identified may directly or indirectly relate to soil disturbance and consequent soil contamination of grass, thereby increasing the rate of exposure of grazing horses to Clostridium botulinum, which resides in soil. Potential relevance: Identification of potentially modifiable risk factors may, ideally following validation in appropriately designed, controlled and randomised intervention studies, lead to practical measures to reduce the incidence of EGS on previously affected premises.

No abstract